Releases: ecmwf/anemoi-core
Releases · ecmwf/anemoi-core
models: v0.5.1
graphs: v0.5.2
training: v0.4.0
0.4.0 (2025-04-16)
⚠ BREAKING CHANGES
- models,training: temporal interpolation (#153)
- config: Improved configuration and data structures (#79)
Features
- Add a CLI to dump the Hydra configuration files into a single YAML file. (#137) (ef1e76e)
- Add EarlyStopping Wrapper (#130) (21d06be)
- Add the possibility to train a model with a dry MLflow run ID (#164) (9849d21)
- config: Improved configuration and data structures (#79) (1f7812b)
- edge post-processor (#199) (1450de7)
- graphs: migrate edge builders to torch-cluster (#56) (f67da66)
- Kcrps (#182) (8bbe898)
- make colormaps configurable for groups of variable (#124) (83d72e1)
- mlflow: Make direct endpoint calls robust (#186) (77bd890)
- models,training: temporal interpolation (#153) (ea644ce)
- models: adding leaky boundings (#256) (426e860)
- training: Add initial TimeLimit callback (#115) (41ff583)
Bug Fixes
- --longtests not available (#200) (9dfec0a)
- Checkpoint path check for multiple tasks/GPUs training (#242) (449f8bd)
- configs,schemas: hierarchical schemas (#221) (2d4a54d)
- correct config comment regarding config_validation flag (#245) (d02e0bb)
- dataset schema too defined too strictly (#143) (4792ee1)
- datashader new release test (#104) (e9c0701)
- dry run forking (#260) (a32cccd)
- graphs,training: skip hidden attributes (#176) (468c45a)
- lam rollout (#213) (6f78387)
- pydantic schemas move (#228) (6bca9bc)
- remove pydantic schemas from checkpoint pickled objects (#237) (ecb945a)
- Rename model_run_ids as trajectory_ids (#216) (e5e942d)
- training,configs: update enc-dec config (#125) (beb8c69)
- training,schema: error when using ReweightedGraphNodeAttribute (#169) (a6313ef)
- training: added weights_only=False flag to torch.load to avoid crashes in torch 2.6 (#205) (02b5117)
- training: Rework Combined Loss (#103) (b63f1aa)
- update stretched.yaml config to be consistent with schema (#195) (21255ac)
- update to null and bump versions (#263) (b4507fb)
- updates to schemas to correct minor pydantic issues (#144) (4e51c17)
Documentation
- Add diagnostics to schema docs (#159) (42c5706)
- Add reference in training getting started guide for intersphinx (#220) (f1d5e1f)
- Add subheadings to schema doc page (#149) (d3c7de9)
- fix documentation to refer to anemoi datasets instead of zarr datasets (#154) (ad062b2)
- models: Docathon (#202) (5dba9d3)
- training: add ADR into docs/ (#168) (ee4da88)
- training: Docathon (#201) (e69430f)
- Update docs for kcrps. (#258) (79cbd1d)
- use new logo (#140) (c269cea)
models: v0.5.0
0.5.0 (2025-04-16)
⚠ BREAKING CHANGES
- models,training: temporal interpolation (#153)
- config: Improved configuration and data structures (#79)
Features
- config: Improved configuration and data structures (#79) (1f7812b)
- Kcrps (#182) (8bbe898)
- models,training: temporal interpolation (#153) (ea644ce)
- models: adding leaky boundings (#256) (426e860)
Bug Fixes
Documentation
graphs: v0.5.1
graphs: v0.5.0
0.5.0 (2025-03-17)
⚠ BREAKING CHANGES
- config: Improved configuration and data structures (#79)
Features
- config: Improved configuration and data structures (#79) (1f7812b)
- graphs: migrate edge builders to torch-cluster (#56) (f67da66)
Bug Fixes
- graphs: make exception agnostic of zarr version (#152) (f26adc9)
- graphs: support torch v2.6 (#122) (dfef377)
Documentation
models: v0.4.2
0.4.2 (2025-02-11)
Features
- make flash attention configurable (#60) (41fcab6)
- models: Copy Imputer (#72) (4690ed5)
- models: normalization layers (#47) (0e1c7c4)
- models: use num_layers of the processor in hierarchical graphs (#78) (7e080ed)
Bug Fixes
- bug in variables ordering in NormalizedReluBounding (#98) (f1cc2e6)
- cancel RTD builds on no change (#97) (36522d8)
- models: 74 imputer inference mode (#127) (0a9cfa7)
- normalise in place to reduce memory (#82) (40dd1a1)
Documentation
training: v0.3.3
0.3.3 (2025-02-05)
Features
- make flash attention configurable (#60) (41fcab6)
- Model Freezing ❄️ (#61) (54e42cf)
- models: normalization layers (#47) (0e1c7c4)
Bug Fixes
- cancel RTD builds on no change (#97) (36522d8)
- only load shards of grid into cpu mem if possible (#83) (abbef4b)
- pin dask version to 2024.12.1 (#94) (074c0f2)
- training: profiler 'Model Summary' works when sharding models over multiple GPUs (#90) (9d9e89a)
- update configs to avoid DeprecationWarning (#53) (3560290)
Documentation
anemoi-graphs: v0.4.4
anemoi-graphs: v0.4.2.post546
0.4.2.post546 (2025-02-04)
What's Changed
Graphs
- feat(graphs,plots): expand support for multi-dimensional node attributes by @JPXKQX in #48
- feat(graphs): New Edge Attribute: AttributeFromNode by @icedoom888 in #62
- feat: support ReducedGaussianGridNodes by @JPXKQX in #54
Other Changes
- pre-commits-for-models-graphs-dev by @sahahner in #45
- ci(docs): bring ReadTheDocs CI pipeline by @JPXKQX in #73
- ci: Reinstantiate CI files by @JesperDramsch in #75
- ci: Propose release-please implementation by @JesperDramsch in #100
Full Changelog: anemoi-graphs-0.4.2...anemoi-graphs-0.4.3